Adopting predictive analytics
The 2008 financial crisis had a crippling effect on businesses operating within the financial sector.
Currently, the pharmaceutical industry is facing similar challenges. As a result, we find pharma has begun to employ multifaceted strategies benchmarking many of the new and emerging methodologies utilized by resilient financial institutions. The foundation of this strategy asks the question, “How does one maintain or gain ‘profitable’ market share without exposing the organization to additional risk?”
This question is pivotal for clinical research sites. Clearly, investigative site profitability is of distinct concern. Site Solutions Summit surveys over the past few years show site profit falling from 20% in 2011 to 13% in 2014.
The answer discovered by most financial institutions was a necessity for strategic partnerships within key demographics, using a vigorous selection and monitoring methodology supported by data analytics. This concept applies to the pharmac industry, as executives are asking a similar question: “How does one profitably expand distribution of new/existing drugs to a broader patient base?” The answer most pharma companies are discovering is similar to that of the financial market—the greatest opportunities are found using vigorous selection and monitoring methodologies based on data analytics to uncover strategically significant global markets. As this logic is applied within the context of business development, strategic partnerships with sites are paramount to achieving alternative revenue streams.
The adoption of predictive analytics as a means to forecast site health and overall quality of service is starting to permeate throughout the pharma industry. One company at the forefront of site analytics is Pfizer, as keynote speaker John Oidtman, vice president of clinical trial support and compliance for Pfizer, outlined at the recent SCOPE conference. Using a variety of tools and technologies, including the use of Saber metrics and Spotfire predictive analytics, his team uncovered statistically significant correlations that predicted clinical trial success related to oversight variables and audits. Their data found that the greater the oversight, the greater the likelihood of trial success and a lower rate of audits.
Through an evaluation of over 6,000 sites, including more than 110,000 patients and the use of “near-time assessments,” Pfizer has developed a multifaceted site selection and oversight strategy. This strategy focuses on two critical aspects related to trial execution: assessing site level trial conduct and assessing the quality of monitoring. With the primary focus on these two attributes, Pfizer has laid the groundwork for an analytics-based mitigation strategy that can be executed uniformly both globally and across individual sites. This targeted approached allowed sites to successfully navigate complex trials by analysis and implementation through:
- Uniformity—A standardized approach to clinical trial management that serves as the foundation for efficient/effective resource deployment targeted at the individual site level
- Reactive Model—A find-and-fix rapid response approach to issue management
- Disparate Data Analysis—Allows Pfizer to analyze scattered data points across many variables on a global and/or site level to identify relationships and trends.
Within this binary framework, Pfizer also has incorporated four predictive algorithms that serve to identify potential strategic partnerships necessary for trial success. Specifically, it is through the application of the following that Pfizer is able to form an overall single numerical metric to determine a clinical site’s health index:
- Historical protocol deviation module
- Site staff turnover rate
- SAE module
- Data timelines module.
While the concept of a site report card is nothing new, the realization sites will be graded and, consequently, included in or excluded from studies based on these and similar formulas can be an intimidating reality. It is clear transparency, quality and accountability of sites is critical to successful completion of cost-effective clinical trials. It is also clear neither sites nor industry can afford to continue a model in which 20% of sites selected perform 80% of the work. Current standards are just not enough to validate the rationale and decision making behind the site selection process. This pressure, coupled with fiscal challenges for both sites and the industry as a whole, creates a new paradigm for us all.
Many initiatives are underway to determine standards and criteria for the report card. It will behoove sites to stay informed, vocalize questions and concerns and provide feedback when given the opportunity during this process, and to take an active role in defining the future of the site selection process.
Jeffrey Adelglass, M.D., F.A.C.S. is founder, owner and president of Research Across America (RAA), a U.S.-based, privately owned, multi-site, multi-discipline clinical research organization. RAA owns multiple research sites across the U.S. and has performed over 1,800 clinical trials in multiple disease areas. Email comments and questions to email@example.com.
This article was reprinted from CWWeekly, a leading clinical research industry newsletter providing expanded analysis on breaking news, study leads, trial results and more. Subscribe »